The objective of this project is to use an established Method of measuring facial activity to determine if negative emotional expressions are related to the occurrence of Transient Myocardial Ischemia (TMI). TMI has been found to be prevelant among patients with coronary atherosclerosis, and a significant predictor of fatal and non-fatal cardiac events. The facial measures have been used previously to differentiate individuals classified as Type A or Type B, and to identify expressions of specific negative emotions found to coincide with changes in cardiac activity. This work will be done in the context of an already funded study where it is possible to examine the relationships between facial expressions of emotion and behavioral interventions designed to reduce the frequency, duration and intensity of TMI episodes. Facial behavior from audio/video records being collected during standard structured interviews will be measured using Ekman and Friesen's Facial Action Coding System. There are three aims: Aim 1: To examine the relationship between expressive behavior and TMI during mental stress tasks. Our purpose is to specify what type of mental stress -- which emotional reaction to stress as evident in the face -- is related to the occurrence of ischemic events. Aim 2: To determine whether the efficacy of the intervention programs designed to reduce TMI is related to the expressive behavior shown prior to intervention, and whether the different interventions influence different aspects of emotional behavior. Aim 3: To explore the relationship between our facial expression measures and the other psychometric measures being collected, in particular to determine how the facial measures of hostility, anger, disgust, and contempt relate to scores on the Cook-Medley Hostility Inventory.